Abstract Stroke is the leading cause of serious long-term disability in the United States. Approximately 40% of patients are left with moderate functional impairments and 15-30% with severe disability. Some degree of spontaneous recovery occurs after an ischemic stroke, but very often this is incomplete. Various neurorestorative therapies, aiming at reducing disability by restoring and reorganizing the surviving brain elements are being tested but none has yet been approved and their optimal prescription remains unknown. At present, the brain changes occurring in the process of spontaneous, or treatment-induced, recovery remain undetermined and there are no available biological markers of recovery. The substrate of clinical recovery is very incompletely understood. It likely includes several different sub-processes, of which vascular and axonal remodeling and the brain's intrinsic ability to repair, are rather prominent. Defining these processes by using non-invasive MRI, laboratory and genetic methods and measurements, if proven valid and reliable, will have the tremendous potential to provide clinicians with tools to rationally guide future medical treatments for stroke recovery. The mains aims of the proposed study are three fold: 1) to define whether specific multiparametric MRI measures, namely cerebral blood volume and blood-to-brain transfer constant values in conjunction with qualitative assessment with susceptibility-weighted imaging, as well as fractional anisotropy, apparent kurtosis coefficient and T1 and T2 values in conjunction with qualitative assessment with q-ball imaging, are capable of characterizing ongoing vascular and axonal remodeling, respectively, in the brain of patients recovering from ischemic stroke, and that these measures and their dynamic temporal evolution correlate with the degree of functional recovery from ischemic stroke at 90 days after stroke onset, 2) to define the levels of vascular endothelial growth factor or stromal derived factor 1 and the numbers or circulating endothelial progenitor cells and whether these levels/numbers and their dynamic temporal evolution correlate with the degree of functional recovery from ischemic stroke at 90 days after stroke onset and with the MRI markers of vascular remodeling, and, 3) to explore whether the presence of specific genotypes, mainly the brain-derived neurotrophic factor Val66Met polymorphism and ApoE subtypes, influence the degree of spontaneous recovery from ischemic stroke and result in attenuated and slower post-stroke vascular and axonal remodeling. Sixty patients will be studied. We expect that the proposed study will produce novel data that will provide new and important information on the morphological brain changes that accompany spontaneous stroke recovery, and will result in the development of an optimal model for rational, image-guided future neurorestorative treatment protocols.